INDEX

Explanations

phrases related to setting something up or preparing something

oai_token-act-pair · gpt-3.5-turbo

phrases related to setting and positioning objects or subjects

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ividual

-0.72

enegger

-0.69

imately

-0.66

entimes

-0.64

 acquired

-0.62

naires

-0.60

inges

-0.60

inia

-0.60

obal

-0.59

naissance

-0.58

POSITIVE LOGITS

abl

1.39

 aside

1.12

 apart

0.96

 precedent

0.81

 loose

0.75

up

0.71

 FIRE

0.69

pH

0.69

 precedence

0.67

flame

0.67

Activations Density 0.096%

phrases related to setting something up or preparing something

phrases related to setting and positioning objects or subjects

No Comments

No Known Activations